拟合中国股票市场收益的统计分布

对中国沪深股市收益的统计分布特征和市场风险规律进行了定量比较研究.分别采用稳定分布、渐近帕累托分布和截断列维分布拟合中国股票市场收益统计分布,实证研究发现中国股市收益分布的中间部分适合用稳定分布描述,分布的尾部适合用尾部指数大于2的渐近帕累托分布描述,即是具有尖峰厚尾特征的有限方差不对称分布.揭示出中国股市中高收益事件比低收益事件发生的更为频繁,深圳市场比上海市场的投资风险要高.所得结论有益于对价格波动性建模、资产定价、金融风险管理等领域的深入研究.

A Study on Describing the Statistical Distribution of Returns in Chinese Stock Markets

The paper quantitatively investigates the statistical characteristics of returns and rules of market risk in Shanghai stock market and Shenzhen stock market.The paper describes the empirical distributions of returns in Chinese stock markets with three theoretical models namely stable distribution,asymptotic Pareto distribution and truncated Levy distribution.By empirical analysis it is found that the centers of empirical distributions can be effectively explained by stable Levy distribution,and the two tails by asymptotic Pareto distribution with characteristic exponents larger than 2.It is indicated that distributions of returns in Chinese stock markets have limited variance and asymmetric heavy tails.The evidences also support that good events happen more frequently than bad losing events in Chinese stock markets,and the market risk of Shenzhen stock market is bigger than one of Shanghai stock market.The conclusions in the paper will benefit the deep research on volatility modeling,asset pricing and financial risk management and so on in future.